E N V I R O N M E N T 40407 D E P A R T M E N T P A P E R S TOWARD ENVIRONMENTALLY AND SOCIALLY SUSTAINABLE DEVELOPMENT P A P E R N o . 1 1 2 MAINSTREAMING ENVIRONMENT IN THE IMPLEMENTATION OF PRSPS IN SUB-SAHARAN AFRICA Sunanda Kishore May 2007 The World Bank Mainstreaming Environment in the Implementation of PRSPs in Sub-Saharan Africa Sunanda Kishore May 2007 Papers in this series are not formal publications of the World Bank. ey are circulated to encourage thought and discussion. e use and cita- tion of this paper should take this into account. e views expressed are those of the authors and should not be attributed to the World Bank. Copies are available from the Environment Department of the World Bank by calling 202-473-3641. TABLE OF CONTENTS ACKNOWLEDGEMENTS....................................................................................................................... v EXECUTIVE SUMMARY.......................................................................................................................vii ABBREVIATIONS AND ACRONYMS................................................................................................. xi CHAPTER 1: INTRODUCTION ............................................................................................................ 1 Poverty Reduction Strategies and their Implementation................................................................ 1 Building on Previous Reviews: Focus on Implementation............................................................. 2 What the Review Covers?.................................................................................................................... 3 PRSP Progress Reports ................................................................................................................ 3 Poverty Reduction Support Credits............................................................................................ 4 Limitations ............................................................................................................................................ 4 CHAPTER 2: FRAMEWORK FOR ASSESSMENT.........................................................................7 Poverty Reduction Strategy Papers.................................................................................................... 7 PRSP Progress Reports........................................................................................................................ 7 Poverty Reduction Support Credits................................................................................................... 7 Scoring................................................................................................................................................... 7 ematic Components of Mainstreaming........................................................................................ 8 Issues............................................................................................................................................... 9 Causal Links Assessment ............................................................................................................. 9 Reponse Systems ......................................................................................................................... 10 Process.......................................................................................................................................... 11 CHAPTER 3: IMPLEMENTATION OF ENVIRONMENTAL PRIORITIES................................ 13 Average Country Scores .................................................................................................................... 13 What has changed in the second PRSP Cycle?........................................................................ 13 Implementation of Environmental Priorities................................................................................. 15 Poverty Reduction Support Credits................................................................................................. 15 Evolving Progression: Country Speci c PRSP Processes.............................................................. 18 Benin PRSP Implementation..................................................................................................... 18 Burkina Faso PRSP Implementation ........................................................................................ 18 Cape Verde PRSP Implementation........................................................................................... 19 Ethiopia PRSP Implementation................................................................................................. 19 Ghana PRSP Implementation.................................................................................................... 20 iii Mainstreaming Environment in the Implementation of PRSPs in Sub-Saharan Africa Madagascar PRSP Implementation .......................................................................................... 21 Mozambique PRSP implementation......................................................................................... 21 Rwanda PRSP Implementation ................................................................................................. 22 Senegal PRSP Implementation.................................................................................................. 23 Tanzania PRSP Implementation................................................................................................ 23 Uganda PRSP Implementation.................................................................................................. 23 CHAPTER 4: CONCLUSIONS AND RECOMMENDATIONS....................................................... 25 REFERENCES........................................................................................................................................... 27 ANNEX 1: Scoring Format..................................................................................................................... 30 ANNEX 2: PRSP implementation progress reports (APRs)............................................................. 31 ANNEX 3: PRSCs and Credit Amount................................................................................................. 32 LIST OF TABLES Table 1. Country-wide distribution of PRSPs, APRs and PRSCs................................................... 3 Table 2. De ning good practice.......................................................................................................... 8 Table 3. Evolving environmental mainstreaming in 11 countries ............................................... 14 Table 4. Implementation progress on the PRSP proposals ........................................................... 15 Table 5. Implementation progress on the PRSCs........................................................................... 16 Table 6. Relative improvement in APRs and PRSCs...................................................................... 17 LIST OF FIGURES Figure 1. Di erences in PRSP I and PRSP II .................................................................................. 14 Figure 2. Environment scores in the context of PRSP implementation...................................... 16 Figure 3. Environment scores in the context of Bank's support through PRSCs to........................17 PRSP I and PRSP II Figure 4. Environment scores in the context of Bank's support through PRSCs to.......................... 17 PRSP implementation Figure 5. Benin PRSP implementation............................................................................................ 18 Figure 6. Burkina Faso PRSP Implementation............................................................................... 19 Figure 7. Cape Verde PRSP Implementation.................................................................................. 19 Figure 8. Ethiopia PRSP Implementation ....................................................................................... 20 Figure 9. Ghana PRSP Implementation........................................................................................... 20 Figure 10. Madagascar PRSP Implementation............................................................................... 21 Figure 11. Mozambique PRSP Implementation............................................................................. 22 Figure 12. Rwanda PRSP Implementation...................................................................................... 22 Figure 13. Senegal PRSP Implementation....................................................................................... 23 Figure 14. Tanzania PRSP Implementation .................................................................................... 23 Figure 15. Uganda PRSP Implementation ...................................................................................... 24 iv Acknowledgments ispaperhasbeenpreparedbySunandaKishore, for their feedback. e author also wishes to ac- consultant, under the guidance of Jan Bojö, World knowledge comments and support from Jan Bojö, Bank Lead Environmental Economist as part of on previous versions of this report. the Environment Department's Work Program. In addition, this paper--the sixth in a series-- e author is grateful to the peer reviewers--Jo- builds on comments from participants at several sep Opio-Odongo, Environment Policy Specialist, seminars both inside and outside the World Bank. UNDP, Vera Wilhelm, Senior Economist, Poverty e author is grateful for their contributions. Sev- Reduction, World Bank and Aziz Bouzaher, Lead eral changes have been made to accommodate Environmental Specialist, Africa Region, World inputs that were received while others remain a Bank for their detailed review of the report and subject of further discussion and investigation. v Executive Summary Environment is recognized as being of central im- e report is based on an assessment of 16 PRSPs portance in the poverty agenda. Planning mecha- in 11 countries, their 25 corresponding PRSP nisms for national development strategies such as Implementation Progress Reports (APRs) and 35 Poverty Reduction Strategy Papers (PRSPs), Me- Poverty Reduction Support Credits (PRSCs). Of dium-Terms Expenditure Frameworks (MTEFs) the 11 African countries considered for this as- and Poverty Reduction Support Credits (PRSCs) sessment, 5 are in their second PRSP cycle. have all made concerted e orts to mainstream environmental issues. As in previous reviews, each of the PRSPs is as- sessed according to 17 variables related to en- is paper aims at presenting a clearer picture of vironmental mainstreaming. e variables are how PRSPs in uence the developmental agenda grouped under four themes: Issues Identi cation, in 11 African countries by assessing the level of Causal Links Assessment, Response Systems, and environmental mainstreaming in the Poverty Re- Process. Issues Identi cation focuses on the di- duction Strategy Process. In this paper, we de ne agnosis of priority environmental concerns and the "PRSP process" as encapsulating the PRSPs, opportunities in PRSPs. Causal Links Assessment their subsequent Progress Reports (APRs)-- analyzes multiple poverty-environment link- which document the extent of implementation of ages. Response Systems covers proposals for en- these strategies, and Poverty Reduction Support vironmental management capacity, investments Credits (PRSCs), a World Bank instrument for in natural resources and human-made capital, concessional support to the implementation of and monitoring of targets and indicators. Finally, PRSPs. As such, this paper (a) assesses the level of Process captures approaches used to promote the environmental mainstreaming across the PRSPs, inclusion of environmental constituencies and an APRs, and PRSCs in 11 African countries; and environmental agenda. Each variable within the (b) provides a detailed desk-review analysis of four themes is ranked on a scale of 0 to 3, and how the implementation of the PRSP in these 11 an unweighted average is calculated on the basis countries is evolving. of these ratings. is average score highlights the extent of environmental mainstreaming achieved e current assessment builds on previously pub- in PRSPs. lished reviews of PRSPs, and it is the sixth report in a series. In the previous assessments, Africa had e APRs and PRSCs are assessed using a reduced constituted the largest sample of Progress Reports version of the 17 variables. As the emphasis of and PRSCs (Bojö et al 2004). e number of Prog- these reports is on "implementation and moni- ress Reports and PRSCs has grown signi cantly toring" rather than diagnosis and analysis, the six since then, and an analysis across all countries aspects/variables speci cally relating to imple- with a PRSP in place was beyond our means. Be- mentation, investments, monitoring, and process cause Sub-Saharan Africa is the World Bank's top are included in the assessment of these reports. priority region, and because we wanted a focus Each variable is scored similarly on a scale of 0 on implementation, a sample of 11 countries was to 3, a er which unweighted average scores are identi ed in Africa, all of which had at least one calculated. APR and at least 2 PRSCs by December 31, 2006. vii Mainstreaming Environment in the Implementation of PRSPs in Sub-Saharan Africa From the previous reviews, it is evident that envi- a marked improvement in environmental ronmental mainstreaming improves signi cantly mainstreaming in their second PRSPs; Gha- as the reports evolve from Interim to Full PRSPs. na, Mozambique and Burkina Faso show a e average score across the sample for environ- marginal decline in average scores in their mental mainstreaming illustrated a huge jump second cycle of PRSPs. from 0.9 in 2002 to 1.3 in 2003 and 1.5 in 2004. Despite an improving average, there exists a large 4. from the cluster of 5 countries in Africa which variation in average scores across sample coun- have moved into their second cycle, causal tries. In these past reviews, several countries are links assessment and the process variable il- identi ed as good practice cases across the four lustrate a big leap in their respective average themes that were assessed--Issues Identi cation, scores (1.8 over 1.4 and 2.6 over 1.8 compared Causal Links Assessment, Response Systems and to the earlier review) while response system Process. has declined to 2 in the current review from 2.3. A larger sample of APRs and PRSCs was included for assessment as a part of the previous review 5. in assessing the integration of environment (Bojö et al 2004). e average mainstreaming there is considerable variability in the average score for these APRs was 1.4 which is close to the score across the 25 PRSP-Progress Reports average score of the entire set of PRSPs (1.5) in (0.7­2.8) and 35 PRSCs (0.7­2.7) across 11 that sample. When comparing progress reports countries. with PRSPs, we found a very weak yet positive correlation between them. is suggests that 6. APRs and PRSCs illustrate a stronger envi- a well mainstreamed PRSP did not necessarily ronmental mainstreaming e ort compared to guarantee a well mainstreamed follow-up on im- the previous review. Also, nine of 11 countries plementation. exemplify a positive leap in their most recent APRs and PRSCs or both (with the exception PRSCs in the previous study, illustrate a high of Madagascar). variability in scores ranging from 0.7 to 2.7, with an average of 1.3 for the sample. In general higher 7. the relationship between PRSPs and APRs is scores (de ned in this case by 1.3­2.7), were usu- evolving to show a positive correlation which ally associated with Sectoral Adjustment Credits is statistically signi cant. is suggests that (SECACs) although with several exceptions. attention to environment in APRs will be in uenced by the level of mainstreaming in PRSP. e main ndings of the current review demon- strate that: 8. PRSCs that respond to rst generation PRSPs illustrate a statistically signi cant weak nega- 1. the assessment of PRSPs, APRs and PRSCs tive correlation while the subsequent PRSCs in the sample establishes a de nite positive that respond to second generation PRSPs il- trend on environmental mainstreaming that lustrate a stronger positive correlation how- is strong and improving. ever, not statistically signi cant. While these numbers can not be taken literally, the trend 2. the PRSP process evolves in a very custom- highlighted from this analysis does support ized and country speci c context across the ndings from the case studies that there is a 11 countries. natural progression to the inclusion of criti- cal environmental priorities in PRSCs. At- 3. while comparing only the 5 with two PRSPs, tention to these priorities usually occurs a er the average score improves slightly from tangible investments in scal reforms, mac- 1.6 to 1.7--Uganda and Tanzania illustrate roeconomic stability and sectoral allocations viii Executive Summary across education and health have been made ty-environment links. While several PRSPs are in a country. common in the sample of the two studies, the conclusion on environmental health can hold 9. while comparing PRSCs to the implemen- for Ghana, Senegal, Rwanda and Ethiopia. For tation of PRSPs through APRs, our analysis understanding other such linkages and the in u- highlights a weak positive correlation (.24). ence that environmental constituents exert on the is illustrates that PRSCs can be better PRSP process, similar analyses are worth consid- aligned to PRSP implementation and vice ering for the future. versa. ird, a well-mainstreamed PRSP may not al- ways guarantee a well-mainstreamed follow-up, e main ndings of the review lead us to certain but the story that is evolving is somewhat opti- conclusions. mistic. Progression in scores and the attention to environment in PRSPs, APRs and PRSCs may not First, the level of environmental mainstream- consistently move in the same direction. How- ing illustrates improvement especially in those ever, a PRSC can be better aligned to PRSP im- countries that have had more time with PRS plementation and an APR is re ecting a stronger implementation and have bene ted from budget mainstreaming e ort on PRSP proposals. While a support directed towards environmental issues. PRSP can suggest several critical priorities, unless While, overall lending has increased for PRSP these priorities are accommodated within a clear countries, relative to non PRSP countries, PRSCs budgetary context, the follow-up in PRS imple- created speci cally to support policy and insti- mentation can not be strengthened to incorpo- tutional reform in PRSP countries account for a rate environmental mainstreaming. larger share of adjustment lending since they are considered to present a better match to the me- Fourth, based on the review, PRSCs that warrant dium-term PRSP programs on account of their the greatest attention to environmental main- longer intended term and more exible modality streaming are the ones well into addressing the of nancial assistance (World Bank 2004). macroeconomic priorities and budgetary reforms in the country, followed by sectoral investments Second, the sample of 5 PRSPs that have moved in health and education. According to a recent into their second cycle considered in this study is independent evaluation on PRSP implementa- too small to make any generalization. e one re- tion (World Bank 2004) annual expenditure as a deeming aspect to increased attention to certain percentage of GDP was higher post-PRSP in the critical aspects in second generation PRSPs is that education and health sectors compared with the it provides some assurance that the PRSP process mean annual expenditure for the pre-PRSP pe- is maturing. is second generation PRSP sam- riod across the 14 countries that were studied. ple has given us an opportunity to revisit causal In a similar analysis of the thematic focus of pre- links assessment and process, which has certain PRSP adjustment lending compared to PRSCs practical implications for PRS implementation. shows that the share of environment and natural e learning from a recent desk review on how resources in PRSCs has grown from 3-6% (World environmental health (one of the weakest causal Bank 2004). However this increase in share for links) is integrated in PRSPs illustrates that in- environment and natural resources is still conser- terventions can be designed and implemented vative compared to public sector governance, hu- to achieve multiple objectives if the underlying man development and social protection. poverty-environment analysis is strong (Kishore 2006). In the current study, we have not focused Fi h, while we can hypothesize on regarding the on assessing whether interventions in better per- extent of mainstreaming in PRS implementation, forming PRSCs or the ones identi ed in APRs the di erence between mainstreaming in a plan are closely tied to a thorough analysis of pover- versus actual implementation is critical. Unfortu- ix Mainstreaming Environment in the Implementation of PRSPs in Sub-Saharan Africa nately, time and resource constraints do not allow Sixth, Ghana, Mozambique, Burkina Faso, Tanza- us to expand the scope of the desk review. What nia, and Uganda seem to be maturing well into the review suggests may be optimistic and a step the PRSP process. ey are well into their sec- up, however from a completely operational per- ond PRSP cycle and have at least two APRs and spective, successful PRSP environmental main- PRSCs which are progressively a ording more at- streaming can decidedly bene t from sectoral tention to environmental concerns. ese coun- strategies being more inclusive and encompass- tries should be considered for country case stud- ing of environmental priorities; and these poli- ies carried out in the eld. cies having a greater buy-in from environmental constituents. x Abbreviations and Acronyms CEA Country Environmental Authority DFID Department for International Development, United Kingdom EDP Environment Department Paper EIA Environmental Impact Assessment HIPC Heavily Indebted Poor Countries IDA International Development Association IMF International Monetary Fund NGO Nongovernmental Organization PREM Poverty Reduction and Economic Management PRSC Poverty Reduction Support Credit PRSP Poverty Reduction Strategy Paper APR Poverty Reduction Strategy Paper Progress Report SAC Structural Adjustment Credit SEA Strategic Environmental Assessment SECAC Sectoral Adjustment Credit UNDP United Nations Development Programme UNEP United Nations Environment Programme WDI World Development Indicators WHO World Health Organization Note: All dollars are U.S. dollars. xi 1 Introduction e importance of environment-poverty links for dium-Terms Expenditure Frameworks (MTEFs) the natural resource, health and vulnerability di- and Poverty Reduction Support Credits (PRSCs) mensions of the livelihoods of the poor is evident have all made concerted e orts to mainstream in development theory and empirical research environmental issues. is process is most ma- (DFID, EC, UNDP and World Bank, 2002). In ture in Sub-Saharan Africa, which explains our line with this thinking, the World Bank's Environ- focus on this region. ment Strategy declared as its goal "...to promote environmental improvements as a fundamental Poverty Reduction Strategies and their element of development and poverty reduction Implementation strategies and actions (World Bank, 2001, p. xx). Poverty reduction strategies (PRS) provide a Unsafe water, inadequate sanitation and waste framework for low-income countries to articulate disposal, and air pollution are a major problem their development priorities and to specify the for the poor--environmental factors are respon- policies, programs, and resources needed to meet sible for almost a quarter of the entire disease bur- their goals. Poverty Reduction Strategy Papers den of developing countries (DFID 2001, WHO (PRSPs)--introduced in 1999 are prepared by 2002). Rapid deforestation and biodiversity losses countries eligible for IDA assistance as a means of are depriving people of valuable forest resources, integrating and strengthening sectoral priorities such as fuel wood, food and medicine. Soil degra- within the context of poverty alleviation. ese dation is a major threat to the livelihoods of 1 bil- strategies are aligned within the larger macro- lion people, mostly the poor who are more likely economic framework for development. PRSPs are to live in degraded or fragile areas. Projections of country-owned and results oriented documents rural population growth and agricultural expan- that provide guidelines and direction for countries sion in the next few decades suggest potentially as they endeavor to utilize foreign assistance for serious pressures that could undermine natural domestic priorities aimed at poverty reduction. resource sustainability in rural areas (Pinstrup- Andersen et al., 1997; Scherr, 1997). But envi- PRSPs prepared in a 3­5 year cycle cover a range ronmental management is not only about protec- of comprehensive strategies focusing on a broad tion, and nding reasonable trade-o s, it is also range of issues--agriculture, industry, mining, about building on the natural resource base to forestry, sheries, water, sanitation, energy, urban foster sustainable growth; as articulated well in and rural development--all of which encapsu- e Wealth of the Poor: Managing Ecosystems to late the need to mainstream environmental con- Fight Poverty (WRI, 2005) and in Our Environ- cerns. As such, they remain central to the policy ment, Our Wealth (UNEP 2006). dialogue between all stakeholders, including the government, donors, private sector, NGO com- As such it is crucial that environment is recog- munity and multilateral development banks in- nized as being of central importance in the pov- cluding the World Bank. e World Bank aims at erty agenda (World Bank 1997). Planning mecha- aligning its lending and non lending activities in nisms for national development strategies such as these countries based on the main priorities iden- Poverty Reduction Strategy Papers (PRSPs), Me- ti ed in PRSPs. 1 Mainstreaming Environment in the Implementation of PRSPs in Sub-Saharan Africa is paper aims at painting a clearer picture of include PRSP-Progress Reports (APRs) and Joint how PRSPs in uence the developmental agenda Sta Assessments (JSAs). e fourth report pub- in 11 African countries by assessing the level of lished in November 2004 expanded on the num- environmental mainstreaming in the Poverty Re- ber of PRSPs to 53, included 21 PRSP Progress duction Strategy Process. In this paper, we de ne Reports (APRs) and as a rst attempt also in- the "PRSP process" as encapsulating the PRSPs cluded a review of 21 PRSCs. In the h review (dra ed in an approximately 3­5 year cycle), their (Kishore 2006), a smaller sample of 16 from the annual subsequent Progress Reports--which existing set of 53 PRSPs was chosen to focus on a document the extent of implementation of these speci c poverty and environment linkage--envi- strategies, and Poverty Reduction Support Cred- ronmental health. e review assessed the extent its, a World Bank instrument for concessional to which environmental health issues and actions support to the implementation of PRSPs in its cli- designed to tackle them were treated in PRSPs. ent countries. As such, this paper (a) assesses the level of environmental mainstreaming across the In the previous assessment, Africa had consti- PRSPs, APRs, and PRSCs in 11 African countries; tuted the largest sample of Progress Reports and and (b) provides a detailed desk-review analysis PRSCs (Bojö et al 2004). e number of Progress of how the implementation of the PRSP in these Reports and PRSCs has grown signi cantly since 11 countries is evolving. then, and an analysis across all counties with a PRSP in place was beyond our means. Because What we mean by "mainstreaming" of the envi- Sub-Saharan Africa is the World Bank's top pri- ronment is summarized here and discussed in ority region, and because we wanted a focus on detail in the methods section below. It is not the implementation, a sample of 11 countries was existence of a stand-alone section or chapter in identi ed in Africa, all of which had at least one the PRSP, nor is it the frequent reference to the APR and at least 2 PRSCs by December 31, 2006. "environment" in a document. e term "main- streaming" is used to denote (a) the description Of the 11 countries, 5 have moved onto their sec- of environmental issues and opportunities; (b) ond PRSPs. e coverage of APRs and PRSCs in the analysis of links between poverty and envi- this review has extended to a total of 60 docu- ronment; (c) the design of responses to meet the ments. is allows us to delve closer into gaug- identi ed challenges; and (d) the inclusion of ing the extent of implementation of PRSPs. e the environmental constituency in the processes methodology used in this study is consistent with leading to the design and implementation of the that used in previous reviews which allows com- PRSP. parisons over time. Building on previous reviews: focus on Each of the PRSPs is assessed according to the 17 implementation variables related to environmental mainstream- ing. e variables are grouped under four themes: e current assessment builds on previously pub- Issues Identi cation, Causal Links Assessment, lished reviews of PRSPs, and it is the sixth report Response Systems and Process. Issues Identi - in a series. A rst assessment of environmental is- cation focuses on the diagnosis of priority envi- sues in the PRSPs of 40 countries was published ronmental concerns and opportunities in PRSPs. as World Bank Environment Department Paper Causal Links Assessment analyzes multiple pov- (EDP) 86 in June 2002. A second EDP (Bojö and erty-environment linkages. Response Systems Reddy 2003a) devoted full attention to the inte- covers proposals for environmental management gration of the targets and indicators related to capacity, investments in natural resources and hu- Millennium Development Goal 7 on environ- man-made capital, and monitoring of targets and mental sustainability. e third report (Bojö and indicators. Finally, Process captures approaches Reddy 2003b) in this series published in Novem- used to promote the inclusion of environmen- ber 2003, extended the review of PRSPs to also tal constituencies and an environmental agenda. 2 Introduction Each variable within the four themes is ranked PRSPs for the current scope of the report. ough on a scale of 1 to 3, and an unweighted average is there may be certain common inclinations in the calculated on the basis of these ratings. is aver- integration of environment as the progression of age score highlights the extent of environmental PRSP implementation continues, it is instructive mainstreaming achieved in PRSPs. to note how each country chalks its own course. Each of these country case studies is discussed in e APRs and PRSCs are assessed using a reduced detail in the later chapters. version of the 17 variables. As the emphasis of these reports is on "implementation and moni- toring" rather than diagnosis and analysis, the six Table 1. Country-wide Distribution of PRSPs, aspects/variables speci cally relating to imple- APRs and PRSCs mentation, investments, monitoring, and process are included in the assessment of these reports. S. No. Country PRSPs APRs PRSCs Total Each variable is scored similarly on a scale of 1 1 Benin 1 1 3 5 to 3, a er which unweighted average scores are 2 Burkina Faso 2 4 6 12 calculated. 3 CapeVerde 1 1 2 4 4 Ethiopia 1 2 2 5 e previous reviews also included a cursory as- 5 Ghana 2 2 4 8 sessment on Joint Sta Assessments (JSAs). ese are not included in the current review, since many 6 Madagascar 1 2 3 6 JSAs for APRs and PRSCs were unavailable for 7 Mozambique 2 3 2 7 review. 8 Rwanda 1 2 2 5 9 Senegal 1 2 2 5 What the Review Covers? 10 Tanzania 2 3 4 9 11 Uganda 2 3 5 10 e report is based on an assessment of 16 PRSPs Total 16 25 35 76 in 11 countries, their 25 corresponding PRSP Implementation Progress Reports and 35 Poverty Reduction Support Credits as seen in Table 1. Of PRSP Progress Reports the 11 African countries considered for this as- sessment, 5 are in their second PRSP cycle1. Since Annual progress reports on the implementa- Africa had the largest sample of PRSP-Implemen- tion of PRSPs highlight e orts of transforming tation Progress Reports and PRSCs in the previ- stated intentions and identi ed priorities in the ous review (Bojö et al 2004), the current review PRSPs into concrete responses or actions. e included only those 11 countries from that sam- guidelines on the PRSP Implementation Progress ple which had at least 1 Implementation Progress Reports presented by the World Bank and IMF Report, and 2 Poverty Reduction Support Credits recommend consistency between national deci- available for review by December 31, 2006. sion making and reporting processes and their integration into annual budget and national de- PRSPs are dra ed in a 3­5 year cycle compared to APRs, which are produced annually. PRSCs encapsulates the annual concessional lending 1 Initially it was expected that PRSPs would be revised and support that the World Bank o ers its client every three years. However many actors have com- countries to PRSPs. For this reason, the country plained that this time frame was too short and that wide distribution of documents in Table 1 shows a an automatic time frame determined externally is dominance of PRSCs and APRs. is dominance seldom aligned with country programming time- of Implementation Progress Reports and PRSCs tables. Following the rst major review of the PRSP inthesamplehelpsestablishtrendsandwarrantsa approach some exibility (up to ve years) in the pe- greater level of attention to the implementation of riodicity of full PRSPs has been introduced . 3 Mainstreaming Environment in the Implementation of PRSPs in Sub-Saharan Africa velopment reports (World Bank and IMF 2004a country in implementing its poverty reduction and 2004b). strategy. ey are becoming an important ve- hicle of IDA nancial assistance, and the roll out In the previous review, of the 14 countries that of PRSCs is gaining momentum. Since the last submitted their 21 corresponding Implementa- review (Bojö et al. 2004)--which included 21 tion Progress Reports, 8 African countries consti- PRSCs of which 11 were identi ed from Africa - tuted the largest sample of 15 APRs. e current the Africa portfolio has grown to 35 PRSCs across review includes all these countries and additions 11 countries (Annex 3). based on documents available Dec 31, 2006 with the exception of Mali, Mauritania and Niger. An- Limitations nex 2 presents a complete list of countries and their implementation progress reports that have e current review is a desk review of PRSP docu- been included in the current review. ments and has the expected set of limitations of similar reviews4. Field-based analysis would add Poverty Reduction Support Credits further depth to this analysis, and will hopefully be possible in the future, at least in a few countries. e World Bank's Poverty Reduction Support Credits (PRSCs), introduced in 2001 are lend- e scale and scope of issues and interventions ing instruments designed to support countries identi ed in these documents need to be viewed eligible for International Development Assistance with caution, as stated intentions may or may not (IDA) in implementing their PRSPs through re ect the real situation. Having said that, the rel- policy and institutional reforms2. PRSCs focus evance of such PRSP reviews should not be mini- on poverty reduction as the central objective of mized--their strength lies in assisting relevant development assistance, while aiming to enhance stakeholders to evaluate if in fact country priori- country ownership, facilitating partnerships with ties and poverty-environmental linkages are be- other institutions, and building on rigorous ana- ing given adequate attention in the implementa- lytical underpinnings for duciary, social, struc- tion of poverty reduction strategies. tural, and sectoral reforms. e development of a PRSC starts with the country's own PRSP and While the framework for this assessment across involves extensive consultations and donor coor- countries has been developed to lend transpar- dination. PRSCs provide customized support to country development and country-owned reform programs3. ey have focused on building gov- 2 See the Operational Memorandum Interim Guide- ernment capacity and institutions, particularly lines for Poverty Reduction Support Credits, May those that serve the poor. 31, 2001, available at http://wbln0011.worldbank. org/Institutional/Manual/OpManual.nsf. e central objective of PRSCs' development as- 3 e new OP 8.60 on Development Policy Lending sistance in achieving poverty reduction is to pro- (DPL) supersedes the old OD 8.60 on Structural Ad- vide customized support to country development justment. All new DPLs, which encompass Poverty and country-owned reform programs. rough Reduction Support Credits are subject to the same PRSCs, the World Bank can support key agricul- requirements--including the determination of any tural, infrastructure, energy, power, manufactur- signi cant e ects on the country's environment, ing, privatization, trade, health, education, and forestry, or other natural resources. e implemen- tation of these requirements will not be discussed water reforms while ensuring transparency, bud- here, but is the subject of a separate report by Mani get discipline and improved nancial manage- and Sears (2006) ment (World Bank 2003). 4 e World Bank and IMF Reviews on PRSP prepa- ration and implementation are also to a large extent A PRSC program usually de nes a series of se- based on the PRSP documents (World Bank and quential operations, which together enable a IMF 2002b, 2003, 2004a, 2004b). 4 Introduction ency, subjectivity in assessments can not be ruled Finally, the documents reviewed in the PRSP out. e aim is not so much scienti c precision process, do not necessarily complete the entire but in identifying and lending credence to the picture of ongoing World Bank e orts. Other underlying trends of environmental integration key documents like the Country Environmental in PRSPs, their Implementation Progress Reports Analysis (CEA), Strategic Environmental As- (APRs) and PRSCs. sessments (SEAs) and the Country Assistance Strategy (CAS), are necessary to obtain a more Another limitation that is shared with preced- comprehensive perspective of the World Bank's ing previous reviews, is the inability of this study involvement in a country as well as understand to account for contextual country di erences. the PRSP implementation process within a larger Country contexts in uence the degree of atten- context. is complete picture must also include tion given to environmental issues and interven- the entire country program, and what other do- tions in PRSPs and their corresponding reports. nors are contributing to relevant domestic e orts. Hence, the scoring should be seen as a screening However, this remains outside the scope of the device, not as the nal judgment on whether or current review. e current review needs to be not a particular issue has been adequately treated, seen as supporting and complementing country given country circumstances. level analyses. 5 2 Framework for Assessment is chapter discusses in greater detail the meth- ports. As in the previous review, this assessment odology used in analyzing the integration of includes also the process variable. is variable environmental concerns across PRSPs, PRSP is included to re ect a transparent participatory Implementation Progress Reports, and PRSCs. mechanism through which environmental con- e assessment framework used in this report stituents are able to voice and include their pri- is built on the previous work on mainstreaming orities and concerns. Although it may be di cult the environment in PRSPs (Bojö and Reddy 2002, to assess the extent of implementation from the 2003a); aligning the environmental priorities of report without a more thorough country-speci c poverty reduction strategies with the Millennium study, the report provides insights into the level of Development Goal on Environmental Sustain- commitment on the part of the government and ability (Bojö and Reddy 2003a); mainstreaming other agencies, and re ects the progress that has the environment in the Country Assistance Strat- been made to date. egies (Ekbom & Bojö 1997; Shyamsundar and Hamilton 2000; Kishore and Shyamsundar 2004; Poverty Reduction Support Credits Kishore 2005); and on the Guidelines for the Joint Sta Assessment of PRSPs (World Bank 2000). Environmental mainstreaming in PRSCs is based on a similar approach used for APRs. In brief, Poverty Reduction Strategy Papers only the variables associated with speci c actions (response system) and process were included. e For the review of PRSPs, ratings have been as- response systems criteria included environmen- signed across 17 variables under four thematic tal management capacity, investment in natural groups to assess environmental integration: (1) capital, investment in human capital, monitoring diagnosis of environmental issues, (2) analysis of natural resource outcomes, and monitoring hu- poverty-environment linkages, (3) environmen- man resource outcomes. e process variable in tally relevant actions and responses, and (4) the the context of PRSCs captures a di erent audience extent to which participation and consultation compared to APRs and PRSPs by paying speci c processes have allowed environmental concerns attention to donor participation, in addition to par- to be included in these strategies. ese 17 vari- ticipation by government environment and natural ables are discussed later in the chapter. resources agencies, and NGO involvement. PRPS Progress Reports Scoring Progress Reports are assessed with respect to the Assessing the 5 new PRSPs (Burkina Faso, Gha- response systems and actions identi ed in the re- na, Mozambique, Tanzania and Uganda) across 7 Mainstreaming Environment in the Implementation of PRSPs in Sub-Saharan Africa 17 variables, and 25 APRs and 35 PRSCs across 6 ough not intended to be scienti cally precise, variables is not practical unless qualitative judg- this scoring method is a practical way to condense ments are formalized and simpli ed. e vari- considerable information into numbers that have ables discussed below are scored with respect to a clear interpretation. e unweighted average each country's documents. A score in the range scores are presented in the results section. We of 0 to 3 is used depending on the treatment of considered applying explicit weights to di erent relevant issues: variables, but this would have made the scoring process less transparent.5 Implicitly, the assign- 0 = no mention ment of the number of variables to each theme 1 = mentioned but not elaborated introduces weighting. 2 = elaborated 3 = good practice Any assessment, including scoring, involves sub- jective judgments. In this format, subjectivity is e rst three scores are related directly to the transparent and consistent across countries. We level of attention given, while the top score implies do not encourage attention to small di erentials a judgment of the quality of the text. is is obvi- in scores between countries. e assessment pro- ously a subjective interpretation, and one that we cess enables us to succinctly present quantita- have tried to illustrate in the sequel by quoting and tive information to complement the qualitative referring to such "good practice" cases. Conceptu- analyses undertaken by the Country Teams and ally, it involves a treatment that is (a) substantive, the PREM Network within the Bank, as well as containing hard data of signi cance; and (b) dis- by external donors and NGOs. An overview of plays a reasoning that is compelling (Table 2). e the aspects incorporated in the scoring format is interpretation of these concepts will vary somewhat presented in the following sections; the scoring depending on what area of scoring we are con- format used is summarized in Annex 1. cerned with. In the case of an issue, a good practice case needs to exhibit such data and reasoning as Thematic Components of Mainstreaming to place the issue solidly on the poverty reduction agenda for that country. In a links-analysis case, e format for assessing environmental priorities the relationship between the variables (e.g. lack of in full PRSPs is grouped into four major themes: safe drinking water and various health indicators) needs to be presented convincingly. In a good prac- · Issues. A description of speci c concerns and tice case of a response, the speci c measures need opportunities relating to the environment. to be de ned, the costs estimated, the institutional · Causal links. An analysis of multiple poverty- responsibility de ned, and a timeline given. For the environment linkages. process variable, one would look for a full descrip- tion of what stakeholder groups were involved, the format and frequency of meetings, the main issues 5 We share our entire scoring sheet, not only the aver- raised, and their follow-up in the PRSP. age score, on request. Table 2: De ning Good Practice Issues Causal Links Assessment Response Systems Process Substantial discussion on and In-depth analysis and of dynamic Assessment of appropriate Signi cant attention to participatory prioritization of the underlying and evolving relationship between regulatory and e ective processes and inclusion of environmental problems poverty and environment including institutional responses, a clear environmental constituencies land tenure, gender, empowerment, identi cation of budgeted macroeconomic distortions etc. interventions which are identi ed across monitorable targets and indicators 8 Framework for Assessment · Responses. An outline of proposals relating answered. First, is poverty contributing to envi- to environmental management, investments ronmental degradation. Second, is environmental in natural and human-made capital, moni- degradation hurting the poor? In this context, we toring, and evaluation. look at seven key linkages to assess the perfor- · Process. Approaches used to promote the in- mance of PRSPs: clusion of environmental constituencies and the environmental agenda. Natural resource degradation and pov- erty. Most poverty is still rural (World Issues Bank, 2002b), and most rural people are directly dependent on the use of natural Priority environmental issues in developing resources to secure a livelihood. Many countries vary signi cantly based on their re- derive a signi cant part of their income source base, problems, and opportunities. Not all directly from non-cultivated resources countries are expected to devote the same level of (Vedeld and others 2004). e linkage attention to all issues. ere are four sub-themes between poverty and the quality of soils, considered: vegetation, and water resources is criti- cal. Land use. Issues relating to soil and sub- soil resources, including mining, erosion, Environmental health. Up to one- h of deserti cation, waterlogging, saliniza- the total burden of disease in the develop- tion, nutrient depletion, and overgrazing; ing world, and close to a third in Sub-Sa- and aboveground resources, including haran Africa, may be associated with en- deforestation and the degradation of for- vironmental risk factors (Lvovsky 2001). ests and woodlands. PRSP analysis of how indoor and out- door pollution, provision of water supply Water. Issues relating to the quantity and and sanitation, and the housing environ- quality of water supply for human con- ment are linked to health outcomes and sumption, irrigation and other uses; wa- the burden of disease can be valuable in ter pollution; coastal zone and marine preparing cost-e ective interventions. aspects; climate variance; and droughts and oods. Vulnerability. Globally, natural hazards claim about 100,000 lives per year, most Air and climate. Issues relating to in- of them in developing countries (DFID door and outdoor pollution--including and others 2002). Analysis of how climate lead, particulate matter, sulfur, nitrogen variability and natural disasters such as oxides, and emissions of greenhouse gas- droughts, oods, earthquakes, and hur- es--from domestic energy use, industrial ricanes impact the poor is valuable for processes, and transport systems. implementing mitigation and adaptation strategies. Biodiversity. Issues relating to the deg- radation of ecosystems, threats to species Property rights. An analysis of how or genetic resources, and opportunities natural resources are "owned" and how for sustainable use. tenure regimes impact their utilization can be of signi cant value. Unequal land Causal links ownership and insecure tenure can force the poor to cultivate marginal environ- Diagnosing a country's environmental issues pro- ments, and may deter long-term invest- vides the foundation for a causal analysis. In such ments (Feder 1987; Heath & Binswanger an analysis, two important questions need to be 1996). 9 Mainstreaming Environment in the Implementation of PRSPs in Sub-Saharan Africa Incentives. Policies relating to pricing, standards, data and information systems, subsidies, taxes, restrictive trade practic- institutional capacity, enforcement ca- es, and the exchange rate can signi cant- pability, and the use of economic instru- ly in uence the use of natural resources ments such as user fees, e uent/emission and the emission of pollutants into the charges, and green taxes (World Bank environment. Gasoline and diesel fuel and IMF 2002a). sales bene ted from about $18 billion in subsidies in 1999, and irrigation from Investment in natural capital. e pro- $10­$15 billion (IMF, UNEP, and World posed programs for natural resource Bank 2002). management indicate the government's priorities and its commitment to improve Empowerment."Empowermentistheex- natural resource productivity. Examples pansion of assets and capabilities of poor include programs supporting the sustain- people to participate in, negotiate with, able management or restoration of soils, in uence, control, and hold accountable forests, woodlands, wetlands, coral reefs, institutions that a ect their lives" (World sheries, and management of protected Bank 2002a, p. vi). In this context, em- areas. powerment concerns the degree to which the poor control decisionmaking regard- Investment in human-made capital. ing a country's resources and environ- Programs relating to slum improvement, ment. While this point is closely related water supply, sanitation, energy e cien- to property rights, it is more concerned cy, waste management, air and water pol- with the level of participation and rules lution, and urban and rural infrastruc- of collective decision making about the ture investments aimed at environmental environment than about legal title. improvements indicate the government's commitment in these areas. Gender and environment. is link draws attention to gender-related policies Monitoring natural resource outcomes. such as the extent to which women have Indicators are important components a voice in the management of commu- of the PRSP monitoring process. In this nal resources, and whether they have the context, targets and indicators for natural right to secure tenure. Women and girls resource management--including land are particularly burdened by the degra- use and soil conservation, such as trends dation of the environment; for example, in productivity or the rate of rehabilita- shortages of fuel wood and water o en tion of degraded lands; forest resources, mean that women must travel longer dis- such as the annual rate of deforestation; tances and spend more time searching area protected, such as the percent of for these resources (OECD 2001). land or sea area protected; water stress or scarcity, such as per capita availability Response systems in cubic meters; and energy, such as de- pendence on traditional energy and the To be meaningful, the discussion on issues and an shi to renewable energy--provide the analysis of causal links must be followed by a set relevant information6. of actions. ese are grouped into ve categories. Environmental management capacity. Environmental management capacity is assessed in terms of actions concerning 6 For detailed discussion on environmental indica- legislation, regulation, environmental tors, see Shyamsundar (2002). 10 Framework for Assessment Monitoring human resource outcomes. ticipatory process is required for identifying and Indicators that measure human resource addressing the concerns of environmental health, outcomes such as health are important. natural resource degradation, vulnerability to nat- Examples include infant mortality and ural disasters, and for undertaking environmental morbidity, such as the infectious and re- investments and monitoring their progress. spiratory disease burden attributable to indoor pollution; access to safe water, It is not possible here to evaluate the quality of such as the percent of the population consultation other than through its expression with access to safe water in rural/urban in the PRSP, APRs and PRSCs. Critics have ar- areas; sanitation, such as the percent of gued that "participatory" events are sometimes population and poor households covered; designed as top-down events, leaving little room and housing standards, such as crowding for upward feedback. It has also been argued that ( oor area/person). consultations o en result in the focus of immedi- ate priorities to the detriment of long-term ones, Process such as those relating to the environment. is may be true in some cases, but the lack of inclu- e description of the process employed in the sion of environmental concerns or actions is then preparation and implementation of a PRSP is part re ected in low ratings under those categories. It of the assessment. Process issues are relevant for should also be recognized that even good faith all aspects of the PRSP, but they are considered consultations sometimes fail to produce a con- in this assessment because an inclusive and par- sensus. 11 3 Implementation of Environmental Priorities is section presents the main results for envi- ally associated with Sectoral Adjustment Credits ronmental mainstreaming in PRSPs, APRs, and (SECACs) although with several exceptions. PRSCs in 11 countries. Given the large number of documents reviewed, the focus is kept on key ndings only. e interested reader is referred to Average country scores the World Bank's internal or external websites, which both contain the full text of the documents Table 3 reports results on the integration of envi- reviewed. ronment in PRSPs for 11 countries. In ve cases, a country moved to its second full PRSP within the From the previous reviews, it is evident that envi- review period. All of them are full PRSPs, as op- ronmental mainstreaming improves signi cantly posed to interim PRSPs, which formed most of the as the reports evolve from Interim to Full PRSPs. rst batch of documents. While our aim is to focus e average score across the sample for environ- on the most mature expression of the PRSP process, mental mainstreaming illustrated a huge jump results from the ve superseded PRSPs are used in from 0.9 in 2002 to 1.3 in 2003 and 1.5 in 2004. the comparison below on the evolution of scores Despite an improving average, there exists a large from PRSP I to PRSP II. While the sample size is variation in average scores across sample coun- small for PRSP II, the average scores between the tries. In these past reviews, several countries are sets of PRSP I and PRSP II remain the same (1.7). identi ed as good practice cases across the four Comparing only the 5 with two PRSPs, the aver- themes that were assessed--Issues Identi cation, age score improved slightly from 1.6. is results Causal Links Assessment, Response Systems and from the fact that Uganda and Tanzania illustrate Process. a marked improvement in environmental main- streaming in their second PRSPs; Ghana, Mozam- A larger sample of APRs and PRSCs was included bique and Burkina Faso show a marginal decline in for assessment as a part of the previous review average scores in their second cycle of PRSPs. (Bojö et al 2004). e average mainstreaming score for APRs was 1.4 which is close to the aver- What has changed in the second PRSP cycle? age score of the entire set of PRSPs (1.5) in that sample. When comparing progress reports with PRSPs moving into their second cycle have the PRSPs, we found a very weak yet positive correla- advantage of building on progress made over pre- tion between them. PRSCs illustrate a high vari- ceding years, aligning a set of new priorities based ability in scores ranging from 0.7 to 2.7, with an on achievements accrued from the implementa- average of 1.3 for the sample. In general higher tion of PRSP I and rede ning the scope of their scores (de ned in this case by 1.3­2.7), were usu- interventions by drawing on past lessons learnt. 13 Mainstreaming Environment in the Implementation of PRSPs in Sub-Saharan Africa Table 3. Evolving environmental mainstreaming in 11 countries S. No. Country Region Overall Score PRSP I Overall Score PRSP II 5 Ghana Africa 2.2 1.9 7 Mozambique Africa 2.2 1.8 3 CapeVerde Africa 1.9 6 Madagascar Africa 1.8 2 Burkina Faso Africa 1.7 1.6 8 Rwanda Africa 1.7 9 Senegal Africa 1.7 4 Ethiopia Africa 1.6 1 Benin Africa 1.5 11 Uganda Africa 1.1 1.7 10 Tanzania Africa 0.9 1.6 Average 1.7 1.7 Looking at the cluster of 5 countries in Africa Available data from World Development Indica- which have moved into their second cycle, this tors 2005 (World Bank 2005) shows that most assumption holds some credence as causal links African countries have a high level of rural popu- assessment (1.8) and the process variable (2.6) lation dependence on natural resources for their illustrate a big leap in average scores (Figure 1). livelihood; high levels of traditional fuel usage; e score for response systems have declined in low levels of access to safe water and adequate the second PRSP cycle. sanitation; and high infant mortality. Low ratings onthe variables identi edunderthethemeCausal While,thesecondgenerationGhanaPRSPbuilds Links Assessment in past reviews have indicated on the main priorities identi ed in PRSP I, the that there is a need for strengthening the analysis attention to time bound, budgeted and monitor- of poverty and environment linkages. is result able interventions identi ed in the second cycle is supported also in a review of the World Bank's is less rigorous compared to PRSP I. is could Country Assistance Strategies (CASs), where pov- very well be in uenced by the premise of limit- erty-environment links are weak (Kishore 2005; ing duplication of information given that on-go- Kishore and Shyamsundar 2004). ing interventions in Ghana are annually tracked under APRs with steadily improving scores - the Ghana APR dra ed earlier the same year as PRSP II in 2005 sustains a high score across response Figure 1. Di erences in PRSP I and PRSP II systems compared to PRSP II (Figure 9 ahead in Thematic Comparison - Five Countries in their Second PRSP Cycle the Chapter). is trend is also illustrated by Mo- 3 zambique (Figure 11 ahead in the Chapter) and 2.5 Burkina Faso (Figure 6 ahead in the Chapter). While Uganda's response system score remains 2 the same across its two PRSPs, Tanzania illus- 1.5 Score trates a marked improved in response systems 1 across PRSP II compared to its very low ranking predecessor. ese results provide some periph- 0.5 eral insights into the shi s across particular as- PSRP I PSRP II 0 Overall Avg Issues Casual Responses Process pects of the PRSP and de nitely warrant further Links investigation. Themes 14 Implementation of Environmental Priorities e average score for Process in the sample for In undertaking a correlation analysis from the PRSP II (2.6) compared to 1.8 for the PRSP I sam- scatter plot (Figure 2) between the PRSP respons- ple provides additional credibility to the assump- es and the PRSP-APRs scores, we nd a positive tion that a higher level of stakeholder involvement and statistically signi cant correlation (0.11). is and participation may be re ected in the second suggests that APRs will invariably encapsulate a cycle of the PRSP process. Following harsh criti- stronger level of environmental integration if the cism in its participation process, an overarching underlying PRSP mainstreaming is strong. review process leading to the formulation of the second cycle of the PRSP process was launched by Poverty Reduction Support Credits the Government of Tanzania in late 2003 to sup- port participation by international NGOs, civil ere is considerable variability in the environ- society organizations, private sector, development mental mainstreaming score across the 35 PRSCs partners and donors, and smaller local NGOs in ranging from a marginal attention (0.7) to consis- identifying cross-cutting issues that strengthen the pro-poor focus of the PRSP (Government of Tanzania 2003). With regard to the levels of op- Table 4. Implementation progress on the PRSP erationalization, Uganda and Tanzania are being proposals regarded as good practice cases for participation PRSP APR in public expenditure reviews supporting sec- S. No. Response Score toral allocations and public expenditure tracking 1. Ghana APR II 2.5 2.8 systems during the implementation of the PRSP (Schnell 2003). Community participation in proj- 2. Ghana APR I 2.5 2.7 ect implementation and service delivery (for ex- 3. Madagascar APR I 2.2 2.7 ample, in water supply and irrigation projects) is 4. Madagascar APR II 2.2 2.5 identi ed as critical in the Uganda and Tanzania 5. Mozambique APR III 3.0 2.5 PRSP II thus boosting their scores for the theme 6. Rwanda APR I 1.7 2.3 Process. 7. Rwanda APR II 1.7 2.3 The implementation of environmental 8. Burkina Faso APR IV 2.2 2.0 priorities 9. Tanzania APR III 1.5 2.0 10. Senegal APR II 2.2 1.8 How a PRSP is implemented is critical. e PRSP 11. Benin APR I 1.8 1.7 Progress Reports and the PRSCs are beginning 12. Burkina Faso APR III 2.5 1.7 to tell us a story in that regard. It is therefore of 13. CapeVerde PRSP- PR I 2.5 1.7 particular interest to see if there is a correlation 14. Ethiopia APR II 2.0 1.7 between the ratings of a PRSP and these subse- 15. Mozambique APR II 3.0 1.7 quent documents. 16. Ethiopia APR I 2.0 1.5 In Table 4, the average score for the progress re- 17. Uganda APR III 1.7 1.5 ports (1.7) lags behind the overall PRSP response 18. Uganda APR II 1.7 1.3 score (2.2). However, the integration of environ- 19. Burkina Faso APR II 2.5 1.2 ment in the implementation of PRSPs as illus- 20. Tanzania APR II 1.5 1.2 trated in the APRs with a score of 1.7 over the 21. Uganda APR I 1.7 1.2 previous review (1.4) is a de nite improvement. 22. Senegal APR I 2.2 1.0 Ghana, Madagascar, Mozambique and Rwanda 23. Tanzania APR I 1.5 0.8 illustrate a steady increase in mainstreaming ef- forts in their Progress reports. Despite this over- 24. Burkina Faso APR I 2.5 0.7 all improvement, there is considerable variation 25. Mozambique APR I 3.0 0.7 within the sample (0.7­2.8). Average 2.2 1.7 15 Mainstreaming Environment in the Implementation of PRSPs in Sub-Saharan Africa Figure 2. Environment scores in the context of Table 5. Implementation progress on the PRSCs PRSP implementation S. No. Country Overall PRSC Score 3.0 1. Benin PRSC II 2.7 2.5 2. Benin PRSC III 2.7 2.0 3. Ghana PRSC IV 2.7 4. Benin PRSC I 2.5 Source 1.5 APR 5. Burkina Faso PRSCVI 2.5 1.0 6. Uganda PRSCV 2.5 0.5 7. Burkina Faso PRSCV 2.3 0.0 1.0 1.5 2.0 2.5 3.0 8. Madagascar PRSC I 2.3 9. Madagascar PRSC II 2.3 PRSP 10. Uganda PRSC IV 2.3 11. Ghana PRSC III 2.2 tent mainstreaming across the themes Response 12. Madagascar PRSC III 2.2 System and Process (2.7). e average score across the sample of 35 PRSCs is 1.8 (Table 5), 13. Rwanda PRSC II 2.2 which illustrates a marked improvement over the 14. Burkina Faso PRSC I 2.0 previous review of PRSCs (1.3). While the sam- 15. Mozambique PRSC I 2.0 ple shows a high variability in the attention that 16. Burkina Faso PRSC II 1.8 environment receives in PRSCs, the fact that 13 17. Burkina Faso PRSC IV 1.8 PRSCs rank higher than 2.2 is noteworthy. 18. Mozambique PRSC II 1.8 19. Uganda PRSC III 1.8 On comparing PRSCs that correspond to the rst round of PRSPs, the correlation is negative and 20. Rwanda PRSC I 1.7 statistically signi cant until PRSCs mature to 21. Tanzania PRSC IV 1.7 respond positively to second generation PRSPs. 22. Ethiopia PRSC II 1.5 However, the positive correlation between PRSCs 23. Tanzania PRSC III 1.5 and second generation PRSPs is not statistically 24. Uganda PRSC II 1.5 signi cant (Figure 3). is provides some cre- 25. Burkina Faso PRSC III 1.3 dence that during the formative years of PRSP 26. Ghana PRSC II 1.3 implementation, focus remains on macroeco- 27. Uganda PRSC I 1.3 nomic and scal reforms, followed by sectoral emphasis on education and health, until the pro- 28. Ethiopia PRSC I 1.2 cess matures to include cross-sectoral priorities 29. Tanzania PRSC II 1.2 including environmental mainstreaming. 30. CapeVerde PRSC II 1.0 31. Senegal PRSC II 0.8 Similarly, as seen in Figure 4, while comparing 32. Tanzania PRSC I 0.8 World Bank's concessional lending to the imple- 33. CapeVerde PRSC I 0.7 mentation of PRSPs, our analysis highlights a pos- 34. Ghana PRSC I 0.7 itive correlation (.24). However, this correlation is not statistically signi cant. is illustrates that 35. Senegal PRSC I 0.7 PRSCs can be better aligned to support environ- Average 1.8 mental mainstreaming across PRSP implementa- tion through APRs and vice versa. e concept of "poverty-reducing expenditures" in countries has emerged from the HIPC expenditure-track- greater budget alignment with the goal of poverty ing exercises. ese expenditures demonstrate reduction (World Bank 2004). 16 Implementation of Environmental Priorities Figure 3. Environment scores in the context of Figure 4. Environment scores in the context Bank's support through PRSCs to PRSPs I and of Bank's support through PRSCs to PRSP PRSP II implementation (APRs) 3.0 3.0 2.5 2.5 2.0 2.0 1.5 1.5 PRSCs PRSCs 1.0 1.0 0.5 0.5 0.0 0.0 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 1 1.5 2 2.5 3 PRSP I APRs 3.0 3.0 2.5 2.5 country which illustrates a weaker integration in environmental mainstreaming across both its 2.0 2.0 most recent APR and PRSC. 1.5 1.5 PRSCs 1.0 1.0 Progression in scores of PRSP, APRs and PRSCs do not consistently move in the same direction. 0.5 0.5 e very "low-rated" Mozambique APR I (refer to 0.0 0.0 gure 11) identi es a scarcity of resources, both 1.0 1 1.2 1.2 1.4 1.4 1.6 1.6 1.8 1.8 2.0 2 2.2 2.2 2.4 2.4 2.6 2.6 2.8 2.8 3.0 3 nancial and human (quali ed technical capac- PRSP II ity across the di erent regions in the country), and limited experience in budgetary execution at the provincial levels as constraints to meeting To illustrate the inconsistency between APRs and PRSCs, an example is that Benin PRSCs foster support to sustainable forest management by in- troducing higher stumpage fees and forestry taxes Table 6. Relative Improvement in APRs and that would re ect the actual value of timber and PRSCs fuelwood in Benin. While forestry sector reform features central in the PRSC, its progress is nei- Country APR PRSC ther highlighted nor discussed in the Benin APR 1 Benin 1.7 0.2 2004. 2 Burkina Faso 1.3 1.8 Table 6 highlights the relative improvement in 3 CapeVerde 1.7 0.3 average scores across the most recent and initial 4 Ethiopia 0.2 0.3 APR and PRSC across the 11 countries. A posi- 5 Ghana 0.1 2.0 tive score suggests a country's constructive and 6 Madagascar ­0.2 ­0.1 a rmative commitment towards environmental 7 Mozambique 1.8 ­0.2 mainstreaming over its earliest APR and PRSC. 8 Rwanda 0.0 0.5 is further insight regarding the implementa- tion status suggests that 9 of 11 countries exem- 9 Senegal 0.8 0.1 plify a leap over the earliest APR and PRSC in the 10 Tanzania 1.2 0.9 inclusion of environment. Madagascar is the only 11 Uganda 0.3 1.2 17 Mainstreaming Environment in the Implementation of PRSPs in Sub-Saharan Africa PRSP objectives and poverty reduction targets Figure 5. Benin PRSP implementation geared towards "highly-rated" environmental 3 outcomes. 2.5 Evolving Progression: Country-Specific 2 PRSP Processes 1.5 Score 1 While a PRSP process is expected to evolve in a customized and country speci c context, in this 0.5 review we seek to identify if a natural maturation 00 pattern exists in the progression of environmental 2002 2003 2004 2005 2006 mainstreaming in PRSP implementation across Year PRSP APR PRSC these 11 countries: Benin, Burkina Faso, Cape Verde, Ethiopia, Ghana, Madagascar, Mozam- bique, Rwanda, Senegal, Tanzania and Uganda. ment. e progress made on PRSCI is followed For each of the 11 countries, an implementa- through in PRSC II and PRSC III. e strategies tion schema has been plotted. ese diagrams focus on rural water supply and sanitation sector illustrate how the environmental mainstreaming reforms; re-enforcement of environmental man- scores are progressing across the PRSP's medium- agement tools, including environmental impact term implementation time-horizon. In several assessment; adoption of a new urban sanitation country cases, the data-points may be very sparse strategy and increasing investments to address (only 4--with the minimal requirement set for urban sanitation and urban air pollution. With this study: 1 PRSP, 1 APR and 2 PRSCs), however regard to forestry, the PRSCs are continuing our intention is to assess if certain common un- support on participatory management of forests derlying trends evolve as the PRSP implementa- and adjacent lands. Given the focus of the lend- tion progresses. ing program on environmental mainstreaming, the Benin PRSCs score relatively high than most Benin PRSP implementation other countries. e PRSP ranks 1.5 on a scale of 1­3, and illus- Burkina Faso PRSP Implementation trates a moderately mainstreamed document. e Progress Report receives a score of 1.7, which is e 2 PRSPs, 4 Annual Progress Reports and the 6 just barely over the PRSP score (Figure 5). e PRSCs illustrate a de nite maturation of the PRSP support provided by the PRSCs to the PRSP im- process. Burkina Faso's PRSP II shows a marginal plementation process illustrates a strong commit- decline in environmental mainstreaming com- ment to environmental integration with scores pared to its earlier PRSP. However, the improving from 2.5 to 2.7.While the APR highlights mea- Progress Report scores (0.7 to 2.0) focus steadily sures undertaken in three main sectors: water on investments in natural resources capital in- supply, sanitation and energy, the progress made cluding sustainable land management, integrated on the implementation of the PRSCs is not cap- soil fertility improvements, upgrading of pastoral tured for forestry in the Progress Report. While lands, and national land management. Progress Reports and PRSCs may not be con- sistent in their reporting across sectors, a scope PRSCs I­VI all focus on environmental main- for stronger alignment of PRSC lending to PRSP streaming but to varying degrees. While continu- implementation (and vice versa) is possible. ing to strengthen environmental management capacity in their support of the PRSP, as PRSCs e focus of the PRSCs continues on rural water, progressed from I to III, the implementation of key health and urban sanitation, and forest manage- reforms became more inclined towards improving 18 Implementation of Environmental Priorities Figure 6. Burkina Faso PRSP Implementation Figure 7. Cape Verde PRSP Implementation 3 3 2.5 2.5 2 2 1.5 1.5 Score Score 1 1 0.5 0.5 0 0 1999 2000 2001 2002 2003 2004 2005 2006 2007 2004 2005 2006 Year Year PRSP APR PRSC PRSP APR PRSC the management of public resources, civil service investments in the key sectors identi ed in the reforms, strengthening basic service delivery in PRSP. the social sectors-health and education, and rais- ing rural incomes by liberalizing the cotton sector. In response to the PRSP, corresponding PRSCs I is led to a decline in the score of PRSCs from I and II support e ectiveness and transparency of to III (Figure 6). As the next set of PRSC opera- the national budget process by promoting public tions (IV­VI) were implemented, environmental expenditure management, civil service reforms mainstreaming components were expanded to and decentralization; developing human capital include energy reforms and electri cation, wa- with a focus on health and education investments ter infrastructure for agricultural diversi cation, and nally, strengthening the sustainability of strengthening environmental impact assessments social protection systems. Environmental main- capacities, and improving access to water and streaming across these reforms has been limited sanitation. Each of these interventions was clearly under the PRSCs as illustrated by low scores aligned with performance indicators leading the across the two PRSCs. Under the scope of pro- PRSC operations to illustrate higher integration of moting good governance, the Ministry of Envi- environmental concerns (1.8 to 2.5). ronment, Agriculture and Fisheries has recently adopted the preliminary sectoral medium term Cape Verde PRSP Implementation expenditure framework. With a score of 1.9, the Cape Verde PRSP illus- Ethiopia PRSP Implementation trates a moderately mainstreamed PRSP (Figure 7). It provides a strong analytical backdrop to Ethiopia's progress on the implementation of the poverty-environmental linkages, and sets forth PRSP illustrates a score of 1.5 across the 6 points responses to strengthen environmental manage- that are assessed, but advances in the APR II to ment capacity by linking them to speci c sectoral 1.7. ese scores are closely in line with the PRSP plans. Investments are directed towards agricul- which scores 1.6 (Figure 8). e Ethiopia Annual tural diversi cation with a focus on soil conser- Progress Report ties its environmental priorities vation, improving irrigation and sound natural to agriculture in response to food insecurity and resources management practices. With regard to vulnerability to natural disasters (in the form of investments in human-made capital, the PRSP severe droughts) is identi ed as a critical issue in identi es electri cation, urban planning, water the PRSP. In addition, APR II also highlights the supply and sanitation as key priority areas. e three environmental proclamations relating to the Annual Progress Report, while brief, responds Environmental Protection Agency and emphasizes adequately to the PRSP by highlighting strategic the importance of agro-processing, land tenure, an 19 Mainstreaming Environment in the Implementation of PRSPs in Sub-Saharan Africa urban management plan focusing on water supply improving environmental mainstreaming scores and sanitation, emergency response systems to di- across the two progress reports and four PRSCs. saster preparedness and mitigation, as responses to However its most recent PRSP II scores slightly PRSP priorities. Despite these initiatives, mecha- lower than PRSP I. nisms for monitoring and evaluating environmen- tal programs especially related to natural resources An issue noteworthy in this review for Ghana has capital are lacking in the progress reports. been that progress under each sector is marked by clearly de ned indicators leading to medium e PRSC operations focus on public expendi- term targets. Also, investments are identi ed un- ture, governance and decentralization, health der HIPC concessional lending, support credits and education, and in reducing vulnerability. and multi-donor budget support. While the outcomes achieved under PRSC I are limited in scope with regard to the environment e overall follow-up on stated intentions of (1.2), moving into PRSC II (which is being dis- PRSP I with proposed actions on environmental bursed as a grant), environmental mainstream- mainstreaming in subsequent PRSCs and Prog- ing is captured through institutional reforms ress Reports is noteworthy in the Ghana PRSP leading to a merger between Ethiopia's Environ- Process as illustrated by increasing score of APRs. mental Protection Agency and Public Enterprise It reinforces the basis for encouraging country Supervising Authority against the nalization owned strategies for pro-poor development. e of the recent Privatization Action Plan; invest- gamut of environmental mainstreaming activi- ments in the horticulture sector under private ties as highlighted in the review, however, ranges sector development initiatives; issuance of the across sectors. rural land proclamations leading to land tenure security; the nalization of a fertilizer plan to Given the signi cant player that agriculture is to support the PRSP's overarching goal of ensuring poverty alleviation, the most recent PRSP II and food security; and nally, adopting a revised ur- the progress on the implementation of PRSP I il- ban land lease to regularize informal settlements lustrate environmental mainstreaming in the ag- in major cities. ricultural sector strategy. Investments in the agri- cultural sector (following the implementation of Ghana PRSP Implementation PRSP I), have been tied to increasing the irriga- tion potential and strengthening land reforms in e review of the PRSP process in Ghana illus- Ghana. Despite improvements in access to water trates that it is maturing given the progressively supply from 2001­2005 (41 percent to 52 per- Figure 8. Ethiopia PRSP Implementation Figure 9. Ghana PRSP Implementation 3 3 2.5 2.5 2 2 1.5 1.5 Score Score 1 1 0.5 0.5 0 0 2002 2003 2004 2005 2003 2004 2005 2006 Year Year PRSP APR PRSC PRSP APR PRSC 20 Implementation of Environmental Priorities cent), access to safe water and sanitation has been Figure 10. Madagascar PRSP Implementation identi ed as crucial throughout the PRSP process 3 against the increasing incidence of guinea worm and rising infant mortality rates. 2.5 2 Although the PRSP score declined in its second 1.5 stage to 1.9, the PRSC scores have shown major Score improvement (1.0 to 2.8) as earlier PRSCs moved 1 beyond the country's macroeconomic priorities 0.5 of strengthening public expenditure and institu- 0 tional performance (Figure 9). A er initializing 2003 2004 2005 2006 sectoral investments in health and education, the Year PRSP APR PRSC evolving Ghana PRSC process highlights a strong commitment towards investments in forestry, strengthening natural resources management and improving water supply and sanitation. porting human development include health, edu- cation, water and social protection. Environment Madagascar PRSP Implementation is mainstreamed through the operationalization of the decentralization strategy, and an active Madagascar PRSP process consistently performs on-going policy dialogue under the PRSC pro- well in integrating environment in its Progress gram supporting Madagascar's third PRSP pillar Reports and PRSCs (Figure 8). e APRs also on broad based growth which includes sheries, stands out as a best practice in providing a de- mining, and rural development. A Public Ex- tailed description of the various environmental penditure Review (PER) which provides speci c policies and plans that have been identi ed and recommendations for the environment sector is a implemented with a medium and long term fo- key achievement for the Madagascar PRSC. cus. ese range from ecotourism development, sheries protection and their sustainable man- Mozambique PRSP Implementation agement, mining with a strict environment re- plenishment and conservation program, disas- ePRSPprocessinMozambiqueshowsastrong ter management, a new forest policy, renewable level of environmental integration and coordina- energy development in rural areas and provision tion across its three progress report, two PRSCs of water supply and sanitation facilities in both and the recent PRSP II. In the recent PRSP II, rural and urban areas. Despite a thorough and environment is mainstreamed across all major exhaustive discussion in APRs, the slight slide sectors in Mozambique: agriculture, health, wa- in score (from 2.7­2.5) is attributable to the fact ter, energy, sheries, and mining. In addition, that actual investments in natural capital and hu- PRSP II extensively discusses the strong envi- man resources capital do not seem as ambitious ronmental health link resulting from a lack of on account of budgetary allocations made in the safe water and poor sanitation. Interestingly, this medium term expenditure review where environ- link in PRSP II is clearly addressed across edu- ment forms only a small part of rural develop- cation, health and water infrastructure. In addi- ment plans (Figure 10). tion, PRSP II, like its predecessor, underscores the importance of early warning systems, and di- e PRSC program for the rst three PRSCs saster preparedness against natural disasters like focuses explicitly on the strategic axes of gover- oods and droughts. Responses to ood control nance and human development. Under gover- and prevention in PRSP II are clubbed again un- nance, the PRSC program includes public nan- der the ambit of water infrastructure develop- cial management, budget support and execution, ment. Despite a strong level of environmental decentralization strategy, and investments sup- integration across sectoral plans, a discussion of 21 Mainstreaming Environment in the Implementation of PRSPs in Sub-Saharan Africa the underlying diagnostic environmental issues across the 17 variables that were reviewed. A is lacking in PRSP II which results in its lower noteworthy aspect of the PRSP was the detailed average score (Figure 11). discussion on the gender di erential and discrim- inatory rights to the access and use of resources. e Implementation Reports continue to build e progress on the PRSP implementation sug- on progress regarding the actions taken on access gests a keen commitment on investments in natu- to land, water, and sanitation; they highlight the ral resources capital and human resource capital measures implemented to guarantee land access including a complete review of existing forestry rights and reduce food vulnerability; describe the legislation, update on soil and water resources actions implemented for ood control, and im- management, promoting the use of manure and/ provements to rural and urban water supply and or chemical fertilizers for increasing agricultural sanitation infrastructure. is leads to progres- productivity, management of water supply sys- sively increasing scores in APRs. Under PRSC I, tems and implementation of sanitation programs, the Government is responsible for undertaking and rural electri cation including fuel substitu- an impact assessment with the objective of iden- tion programs. tifying linkages between key macroeconomic and structural reforms policies and the environment For Rwanda the second operation of the PRSC in an e ort to mainstream the environmental sus- program, is being implemented as a Poverty Re- tainability into the subsequent PRSC process. e duction Support Grant (PRSG). PRSG II builds implementation of natural resources management on the progress underway from PRSC I. In addi- plans across several communities are being imple- tion to creating a climate that promotes macro- mented under PRSC I and II. PRSC II also focuses economic stability and a favorable private sector on establishing operational and monitoring sys- investment climate, the PRSG focuses on improv- tems for agriculture and land productivity. Details ing the coverage of service delivery through im- regarding these plans are minimal in PRSC II. proved expenditure e ciency, and expands the is invariably impacts the score for PRSC II (Fig- mandate of PRSC I from health and education to ure 11). Water supply and sanitation programs are include access to water supply and sanitation. Pre- expected to be implemented under PRSC III. pared in consultation with districts and provinces, a Strategic Plan for Agricultural Transformation Rwanda PRSP Implementation (PSTA) remains a key milestone under the PRSG process which guarantees progress on the imple- Rwanda PRSP highlights a moderate level of en- mentation of land reforms (a critical and sensi- vironmental mainstreaming with a score of 1.7 tive issue for Rwanda), establishment of Rwanda Figure 11. Mozambique PRSP Implementation Figure 12. Rwanda PRSP Implementation 3 3 2.5 2.5 2 2 1.5 1.5 Score Score 1 1 0.5 0.5 0 0 2001 2002 2003 2004 2005 2006 2002 2003 2004 2005 Year Year PRSP APR PRSC PRSP APR PRSC 22 Implementation of Environmental Priorities Environmental Management Authority (REMA) Figure 14. Tanzania PRSP Implementation and the completion of EIA guidelines. is pro- 3 vides a strong basis for allocating the PRSG with 2.5 a higher score over PRSC I (Figure 12). 2 Senegal PRSP Implementation 1.5 Score 1 Environmental mainstreaming across the Sen- egal PRSP process remains relatively weak given 0.5 the low scores across the PRSCs and APRs. APR 0 II which rated 1.7 however does report on some 2001 2002 2003 2004 2005 2006 Year progress relating to investments in natural re- PRSP APR PRSC sources and man-made capital (Figure 13). While the integration of environmental concerns is weak, it seems in line with a common trend that is evident in PRSCs--they seem to move towards ment. Compared to PRSP I, however investments focusing on investments that strengthen pov- identi ed in PRSP II on natural resource capital erty-environmental linkages and environmental and human-made capital are weak. management capacity only once the main social sectors--education and health--and macroeco- e integration of environment in the three nomic priorities have been addressed. implementation progress reports has improved steadily over time. is same trend is also illus- Tanzania PRSP Implementation trated in the Poverty Reduction Support Cred- its (I­IV) as the score improves from 0.8 to 1.7 Environmental mainstreaming in PRSP II has (Figure 14). e focus of investments identi ed improved considerably as illustrated by a score in the PRSCs is on water supply and sanitation of 1.6 over PRSP I, which ranked 0.9. e focus programs, and on strengthening institutional and of environmental mainstreaming in the PRSP II regulatory capacity in the country. Investments in has been on poverty-environment linkages, envi- natural resources capital are weak in the PRSCs as ronmental health, strengthening environmental well as the Progress reports. management capacity and more speci cally on improving the voice of the environmental con- Uganda PRSP Implementation stituency in Tanzania as re ected in the docu- e Uganda PRSP process with two PRSPs, three Implementation Progress Reports and 5 Poverty Reduction Support Credits is well into its matu- ration state. e most recent Uganda Progress Figure 13. Senegal PRSP Implementation Report receives an average score of 1.5 across the 3 6 points that are assessed (Figure 15). Although 2.5 the progress reported on the implementation of environmental proposals in not completely sat- 2 isfactory, it is still an improvement over the pre- 1.5 Score vious progress reports. However, investments 1 drawn out for water supply and sanitation are noteworthy and stand as a good practice for oth- 0.5 er countries. Progress remains weak in the im- 0 2002 2003 2004 2005 2006 plementation process relating to natural resource Year capital as well as identifying stakeholders for and PRSP APR PRSC bene ciaries of the proposed environmental ac- 23 Mainstreaming Environment in the Implementation of PRSPs in Sub-Saharan Africa Figure 15. Uganda PRSP Implementation approved by the Cabinet in November 2004 . It builds on the fourth PRSC operation which was 3 disbursed as Poverty Reduction Support Grant. 2.5 While the PRSC operations have focused to vary- 2 ing degrees on strengthening environmental management capacity, overtime they expanded 1.5 Score in scope to support programs related to natural 1 resources and human made capital. is included 0.5 increasing land security and titling; consolidat- ing environmental impact assessment guidelines 0 2000 2001 2002 2003 2004 2005 across several sectors (roads, water supply and Year sanitation; energy; mining, agriculture, forestry, PRSP APR PRSC and wildlife management), improving forest re- sources management and water resource manage- ment; increasing access to water and sanitation system, and improving water facilities mainte- tivities. is aspect--particularly relating to the nance. Since, each of these interventions identi- inclusion of an environmental constituency in ed in PRSC V is accompanied by sector speci c the PRSP process--has improved in Uganda's medium term monitorable targets, it allows the PRSP II (1.7). allocation of a higher score particularly across the two monitoring variables under the theme Building on four preceding PRSCs, the recent Responses leading to an overall score of 2.5 for PRSC V is the rst to support Uganda's PRSP II PRSC V. 24 4 Conclusions and Recommendations e main ndings of the review lead us to certain 2006). In the current study, we have not focused conclusions. on assessing whether interventions in better per- forming PRSCs or the ones identi ed in APRs First, the level of environmental mainstream- are closely tied to a thorough analysis of pover- ing illustrates improvement especially in those ty-environment links. While several PRSPs are countries that have had more time with PRS common in the sample of the two studies, the implementation and have bene ted from budget conclusion on environmental health can hold support directed towards environmental issues. for Ghana, Senegal, Rwanda and Ethiopia. For While, overall lending has increased for PRSP understanding other such linkages and the in u- countries, relative to non PRSP countries, PRSCs ence that environmental constituents exert on the created speci cally to support policy and insti- PRSP process, similar analyses are worth consid- tutional reform in PRSP countries account for a ering for the future. larger share of adjustment lending since they are considered to present a better match to the me- ird, a well-mainstreamed PRSP may not al- dium-term PRSP programs on account of their ways guarantee a well-mainstreamed follow-up, longer intended term and more exible modality but the story that is evolving is somewhat opti- of nancial assistance (World Bank 2004). mistic. Progression in scores and the attention to environment in PRSPs, APRs and PRSCs may not Second, the sample of 5 PRSPs that have moved consistently move in the same direction. How- into their second cycle considered in this study is ever, a PRSC can be better aligned to PRSP im- too small to make any generalization. e one re- plementation and an APR is re ecting a stronger deeming aspect to increased attention to certain mainstreaming e ort on PRSP proposals. While a critical aspects in second generation PRSPs is that PRSP can suggest several critical priorities, unless it provides some assurance that the PRSP process these priorities are accommodated within a clear is maturing. is second generation PRSP sam- budgetary context, the follow-up in PRSP imple- ple has given us an opportunity to revisit causal mentation can not be strengthened to incorporate links assessment and process, which has certain environmental mainstreaming. practical implications for PRSP implementation. e learning from a recent desk review on how Fourth, based on the review, PRSCs that warrant environmental health (one of the weakest causal the greatest attention to environmental main- links) is integrated in PRSPs illustrates that in- streaming are the ones well into addressing the terventions can be designed and implemented macroeconomic priorities and budgetary reforms to achieve multiple objectives if the underlying in the country, followed by sectoral investments poverty-environment analysis is strong (Kishore in health and education. According to a recent 25 Mainstreaming Environment in the Implementation of PRSPs in Sub-Saharan Africa independent evaluation on PRSP implementa- nately, time and resource constraints do not allow tion (World Bank 2004) annual expenditure as a us to expand the scope of the desk review. What percentage of GDP was higher post-PRSP in the the review suggests may be optimistic and a step education and health sectors compared with the up, however from a completely operational per- mean annual expenditure for the pre-PRSP pe- spective, successful PRSP environmental main- riod across the 14 countries that were studied. streaming can decidedly bene t from sectoral In a similar analysis of the thematic focus of pre- strategies being more inclusive and encompass- PRSP adjustment lending compared to PRSCs ing of environmental priorities; and these poli- shows that the share of environment and natural cies having a greater buy-in from environmental resources in PRSCs has grown from 3­6% (World constituents. Bank 2004). However this increase in share for environment and natural resources is still conser- Sixth, Ghana, Mozambique, Burkina Faso, Tanza- vative compared to public sector governance, hu- nia, and Uganda seem to be maturing well into man development and social protection. the PRSP process. ey are well into their second PRSP cycle and have at least two of PRSP-APRs Fi h, while we can hypothesize on regarding the and PRSCs which are progressively a ording extent of mainstreaming in PRS implementation, more attention to environmental concerns. ese the di erence between mainstreaming in a plan countries should be considered for country case versus actual implementation is critical. Unfortu- studies carried out in the eld. 26 References Bojö, Jan, and Rama Chandra Reddy. 2002. "Pov- Feder, G. 1987. "Land Ownership, Productivity erty Reduction Strategies and Environment: and Farm Productivity: Evidence from Ru- A Review of 40 Interim and Full Poverty ral ailand." Journal of Development Studies Reduction Strategy Papers." Environment 24(1): 16­30. Department Paper No. 86, Environmental Government of Tanzania. 2003. "Guide for Pov- Economics Series. World Bank, Washington, erty Reduction Strategy Review". Vice Pres- D.C. ident's O ce. < http://www.tz.undp.org/ Bojö, Jan, and Rama Chandra Reddy. 2003a. Guidelines_PRSReview.doc> Dated: 1 March "Poverty Reduction Strategies and Millen- 2007 nium Development Goal on Environmental Heath, J. and H. Binswanger, 1996. "Natural Re- Sustainability: Opportunities for Alignment." sources Degradation E ects of Poverty and Environment Department Paper No. 92. En- Population Growth Are Largely Policy In- vironmental Economics Series. World Bank, duced: e Case of Colombia." Environment Washington, D.C. and Development Economics 1 (Part 1): 65­ Bojö, Jan, and Rama Chandra Reddy. 2003b. "Sta- 83. tus and Evolution of Environmental Priorities International Monetary Fund (IMF), United Na- in the Poverty Reduction Strategies: An As- tions Environment Programme (UNEP), and sessment of Fi y Poverty Reduction Strategy the World Bank. 2002. Financing Sustainable Papers." Environment Department Paper No. Development. Washington, D.C. 93, Environmental Economics Series. World Kishore, S. 2005. Environment in 2005 Country Bank, Washington, D.C. Assistance Strategies. Environment Depart- DFID. 2001. Poverty and the Environment: What ment Paper No.110, Environmental Econom- the Poor Say; An Assessment of Poverty-En- ics Series. Environment Department, World vironment Linkages in Participatory Poverty Bank, Washington D.C. Assessments, Environment Policy Depart- Kishore, S. 2006. Environmental Health Issues in ment, Issues Paper No. 1, October 2001, Poverty Reduction Strategies: A Review. En- Department for International Development: vironment Strategy Paper No. 12. Environ- London ment Department, World Bank, Washington DFID, EC, UNDP and World Bank, 2002. Link- D.C. ing Poverty Reduction and Environmental Kishore, S. and Shyamsundar, P. 2004. An Envi- Management: Policy Challenges and Oppor- ronmental Review of 2002-04 Country Assis- tunities. Published by the World Bank, Wash- tance Strategies. Environmental Economics ington, D.C. Series, Environment Department Paper No. Ekbom, A. and J. Bojö. 1997. "Mainstreaming En- 105. Washington D.C. vironment in Country Assistance Strategies." Lvovsky, K. 2001. "Health and Environment." Discussion Paper No.1. World Bank, Envi- Environment Strategy Papers. Strategy Series ronment Group, Africa Region, Washington No.1. World Bank, Washington, D.C. D.C. 27 Mainstreaming Environment in the Implementation of PRSPs in Sub-Saharan Africa Mani, M. and Sears, A. 2006. Implementation of ------. 2001. Making Sustainable Commitments: Environment and Natural Aspects in Devel- An Environment Strategy for the World Bank. opment Policy Lending. Environment De- World Bank, Washington, D.C. partment Paper No. 111. World Bank, Wash- ------. 2002a. Empowerment and Poverty Re- ington D.C. duction: A Sourcebook. PREM. Washington, OECD. 2001. "Poverty-Environment-Gender D.C. Linkages." O -print of the DAC Journal 2001 ------. 2002b. Reaching the Rural Poor: A Re- 1 (4). newed Strategy for Rural Development. World Pinstrup-Andersen, P. 1997. e World Food Situ- Bank, Washington, D.C. ation: Recent Developments, Emerging Issues, ------. 2003. Putting Our Commitments to Work: and Long-Term Prospects. IFPRI, Washing- Environment Strategy Implementation Prog- ton, DC ress Report. World Bank. Washington, D.C. Scherr, S.J. 1995. Economic analysis of agroforestry ------. 2004. e Poverty Reduction Strategy systems: the farmers' perspective. In: Current, Initiative: An Independent Evaluation of e D., Lutz, E. & Scherr, S. (eds.) 1995. Costs, World Bank's Support rough 2003. World bene ts and farmers' adoption of agroforest- Bank, Washington DC. ry: project experience in Latin America and ------. 2005. World Development Indicators 2005. the Caribbean. World Bank, Washington, World Bank, Washington, D.C. D.C. World Bank and International Monetary Fund Schnell, S. 2003 "Participation in Monitoring and (IMF). 2002a. A Sourcebook for Poverty Re- Evaluation of PRSPs: A Document Review of duction Strategies. Volumes I and II. Wash- Trends and Approaches Emerging from 21 Full ington, D.C. PRSPs." World Bank, Washington DC ------.2002b.GuidelinesforWorldBankandIMF Shyamsundar, P. 2002. "Poverty-Environment Sta s for Joint Sta Advisory Notes (JSANs of Indicators." Environment Department Paper PRS Annual Progress Reports (APR) on Imple- No. 84. World Bank, Washington, D.C. mentation. World Bank, Washington, D.C. Shyamsundar, P., and K. Hamilton. 2000. "An En- ------. 2003. "Poverty Reduction Strategy Pa- vironmental Review of 1999 Country Assis- per--Detailed Analysis of Progress in Im- tance Strategies--Best Practice And Lessons plementation." Prepared by the sta s of the Learned." World Bank, Environment Depart- World Bank and IMF, Washington, D.C. ment, Environmental Economics and Indica- ------. 2004a. Guidelines for World Bank and IMF tors Unit, Washington, D.C. Processed. Sta s for Joint Sta Advisory Notes (JSANs of UNEP. 2006. Our Environment, Our Wealth. Af- PRS Annual Progress Reports (APR) on Imple- rica Environment Outlook 2, United Nations mentation. World Bank, Washington, D.C. Environment Program, Nairobi. ------. 2004b. Poverty Reduction Strategy Papers Vedeld, P. A. Angelsen, E. Sjaastad, and G. Ko- (PRSP)--Progress in Implementation. Pre- bugabe-Berg. 2004 "Counting on the Envi- pared by the sta s of the World Bank and ronment: Forest Income and the Rural Poor." IMF. World Bank, Washington, D.C. Environment Department Paper No. 09. World Health Organization (WHO). 2002. World World Bank, Washington, D.C. Health Report 2002: Reducing Risks, Promot- World Bank. 1997. Can the Environment Wait: ing Healthy Life. WHO, Geneva, Switzerland. Priorities for East Asia. World Bank, Wash- ington, D.C. 28 Annexes 29 Mainstreaming Environment in the Implementation of PRSPs in Sub-Saharan Africa Annex 1 Scoring Format of the PRSP Assessment 1. Issues in Focus 1. Land use: degradation, deforestation, erosion, overgrazing, etc. 2. Water: drinking water, irrigation, sheries and water pollution, etc. 3. Air & climate: air quality, solid fuel usage, emissions, climate variability 4. Biodiversity: threats to ecosystems, species and genes, nature-based opportunities 2. Causal Link Assessment 1. Poverty and NR degradation: resource dependence and inequality 2. Environmental health: water and air pollution related to disease 3. Vulnerability: impacts of natural hazards 4. Property rights: tenure and user rights 5. Incentives: pricing interventions, taxation, subsidies, exchange rate, trade, etc. 6. Empowerment: community-based management, decentralization, and partnerships 7. Gender: role of women in environmental management 3. Response systems 1. Environmental management capacity: legislation, regulation, institutional reform, data systems, cross-sectoral coordination, , environmental stan- dards, environmental economic instruments, etc 2. Investment in natural capital: investment in sustainable natural resource management, e.g. watershed management 3. Investment in human-made capital: investment in environmental infrastructure, e.g. sewage treatment plants 4. Monitoring natural resource outcomes: deforestation, a orestation, rehabilitated areas, protected areas, soil & water conservation measures, renew- able energy use, etc. 5. Monitoring human resource outcomes: infant and child mortality, disease burden related to environmental risk factors, time spent collecting fuelwood and water 4. Process 1. Description of the participatory process and inclusion of environmental constituencies, particularly with respect to the identi cation of environmental issues, poverty links, and actions Score: 0 = not mentioned 1 = mentioned but not elaborated 2 = elaborated 3 = good practice 30 Annexes Annex 2 PRSP implementation progress reports (APRs) Country APR Date 1 Benin Benin APR I December 2004 2 Burkina Faso Burkina Faso APR I September 30, 2001 3 Burkina Faso APR II September 30, 2002 4 Burkina Faso APR III December 31, 2003 5 Burkina Faso APR IV December, 2004 6 CapeVerde CapeVerde PRSP- PR I May 2006 7 Ethiopia Ethiopia APR I December 31, 2003 8 Ethiopia APR II March 2005 9 Ghana Ghana APR I March 31, 2004 10 Ghana APR II March, 2005 11 Madagascar Madagascar APR I July 31, 2004 12 Madagascar APR II June, 2006 13 Mozambique Mozambique APR I February 28, 2003 14 Mozambique APR II March 2004 15 Mozambique APR III June 7, 2005 16 Rwanda Rwanda APR I June 30, 2003 17 Rwanda APR II October 31, 2004 18 Senegal Senegal APR I March 31 2004 19 Senegal APR II May, 2005 20 Tanzania Tanzania APR I August 14, 2001 21 Tanzania APR II March 31, 2003 22 Tanzania APR III April 30, 2004 23 Uganda Uganda APR I March 2, 2001 24 Uganda APR II March 31, 2002 25 Uganda APR III August 13, 2003 31 Mainstreaming Environment in the Implementation of PRSPs in Sub-Saharan Africa Annex 3 PRSCs and Credit Amount Country Credit Amount (in millions) Date 1 Benin PRSC I 20 February 23, 2004 2 Benin PRSC II 30 May 16, 2005 3 Benin PRSC III 30 October 2, 2006 4 Burkina Faso PRSC I 45 July 26 2001 5 Burkina Faso PRSC II 35 June 20, 2002 6 Burkina Faso PRSC III 50 March 16 2003 7 Burkina Faso PRSC IV 50 April 14 2004 8 Burkina Faso PRSCV 60 April 7, 2005 9 Burkina Faso PRSCVI 60 May 16, 2006 10 CapeVerde PRSC I 15 January 26, 2005 11 CapeVerde PRSC II 10 November 10, 2006 12 Ethiopia PRSC I 120 May 15, 2002 13 Ethiopia PRSC II 130 November 1, 2004 14 Ghana PRSC I 125 May 29, 2003 15 Ghana PRSC II 125 June 7, 2004 16 Ghana PRSC III 125 July 25, 2005 17 Ghana PRSC IV 140 May 1, 2006 18 Madagascar PRSC I 125 June 23, 2004 19 Madagascar PRSC II 80 June 9, 2005 20 Madagascar PRSC III 40 June 14, 2006 21 Mozambique PRSC I 60 June 9, 2004 22 Mozambique PRSC II 120 August 4, 2005 23 Rwanda PRSC I 65 June 23, 2004 24 Rwanda PRSC II 55 October 13, 2005 25 Senegal PRSC I 30 November 3, 2005 26 Senegal PRSC II 30 February 28, 2006 27 Tanzania PRSC I 100 April 29, 2003 28 Tanzania PRSC II 90 July 6, 2004 29 Tanzania PRSC III 150 August 10, 2005 30 Tanzania PRSC IV 200 April 10, 2006 31 Uganda PRSC I 65 March 23, 2001 32 Uganda PRSC II 55 May 31, 2002 33 Uganda PRSC III 55 July 28, 2003 34 Uganda PRSC IV 15 August 5, 2004 35 Uganda PRSCV 135 December 13, 2005 Total 2640 32 THE WORLD BANK 1818 H Street, NW Washington, DC 20433 USA Telephone: 202-473-1000 Facsimile: 202-477-6391 Internet: www.worldbank.org E-mail: feedback@worldbank.org